KR101812682B1 - Dynamic geo-location parameter for determining an impact of online behavior on offline sales - Google Patents

Abstract

The online behavior of the users associated with the product is captured. In addition, offline sales for the product are captured. Geolocation parameters are also determined for the captured on-line and off-line data. The geographic location parameters for the product and the captured data are used to determine a minimum geographic location parameter with statistical significance for estimating the effect of online behavior on offline sales.

Online behavior, offline sales, minimum geolocation parameters

Description

[0002] DYNAMIC GEO-LOCATION PARAMETERS FOR DETERMINING AN IMPACT OF ONLINE BEHAVIOR ON OFFLINE SALES [0002]

[preference]

This application claims priority to US Provisional Patent Application No. 61 / 105,674, filed October 15, 2008, entitled " Dynamic Regioning for Online and Offline Cross Optimization Parameters " U.S. Patent Application No. 12 / 366,118 filed on May 5, the entirety of which is incorporated by reference in its entirety.

In today's digital age, companies selling goods and services to their customers are You must participate in online marketing and sales. For example, many large department stores, which traditionally include brick and mortar stores, also have sophisticated websites that provide detailed product information and features that enable visitors to purchase products online. to provide. In addition, many companies are making mass marketing budgets for online marketing, including marketing on their websites.

One of the major problems these companies face is how they will evaluate their online marketing efforts. Online activities such as website traffic and online sales can be used as a measure of online marketing efforts. However, online marketing can affect online sales as well as in-store sales. For example, a customer can view product information online and then go to a traditional store to view the product and ultimately purchase the product from the store. It is very difficult to track the impact of online marketing when purchasing is done in this way. Marketing efforts To optimize and provide the basis for spending on online marketing, companies need to have the ability to accurately capture the impact their online marketing efforts have on off-line sales.

An object of the present invention is to solve all of the problems of the prior art as described above.

Another object of the present invention is to provide a dynamic geolocation parameter for determining the effect of online behavior on offline sales.

A typical configuration of the present invention is as follows.

According to an aspect of the present invention there is provided a computer system configured to determine a geographic location parameter for a product, the computer system comprising: a data storage device for storing online behavior data and offline sales data for the product; A computer system comprising a dynamic localization module for determining a minimum geographic location parameter having statistical significance for estimating its impact and an integration module for integrating the stored online behavior data and offline sales data corresponding to the minimum geographic location parameter, / RTI >

According to another aspect of the present invention there is provided a computer system configured to correlate an online activity with an offline sale, the computer system comprising: a network interface that receives online activity data captured from web sites and receives offline sales data captured from a store sale; A data storage device for storing received online behavior data and offline sales data, and a processor, wherein the processor is configured to determine an online behavior of users associated with the product, determine geographic location parameters of the users' And determining a minimum geographic location parameter having statistical significance for estimating the effect of the online action on offline sales of the product using the online behavior and the geographic location parameters.

According to another aspect of the present invention there is provided a method for determining a minimum geographic location parameter having statistical significance for estimating the effect of online behavior on offline sales when executed on one or more computer systems, A computer readable medium, comprising: determining an online behavior of users associated with a product; determining geographic location parameters for users' online behavior; and determining, using the online behavior and geographic location parameters, There is provided at least one computer readable medium comprising a minimum geographic location parameter having statistical significance for estimating the effect of an online act on offline sales of the product.

Embodiments of the present invention will be described in detail below with reference to the following drawings.

According to the present invention, the above-described problems of the prior art are solved.

In accordance with the present invention, dynamic geolocation parameters are provided to determine the effect of online behavior on offline sales.

For simplicity and illustrative purposes, the principles of the embodiments are described primarily with reference to examples. In the following description, numerous specific details are set forth in order to provide a clear understanding of the embodiments. However, it will be apparent to those skilled in the art that the embodiments may be practiced without being limited to these specific details. In some instances, well-known methods and configurations are not described in detail in order not to unnecessarily obscure the embodiments.

1. Overview

The online behavior of the users associated with the product is captured. The online action may include any measurable or traceable event of the user on the Internet. This may include visits to a website, frequency of visits to the page, monitoring what is clicked on the website, and so on. The geolocation parameters for users are also determined. The geolocation parameter is the geographic location for the user to capture the online action. In one example, the geolocation parameter for the user is determined using a reverse Internet Protocol (IP) lookup. This includes determining a user's IP address and identifying a geographical location for that IP address, e.g., a zip code, city, or some other location.

The user or user group may have multiple geographic location parameters. These multiple geographic location parameters are referred to as granularities. For example, the reverse IP address lookup may identify a minimum granularity geolocation parameter for a user, such as a zip code. Other higher The granularities can be determined from the zip code. For example, an area can combine multiple neighborhood zip codes. The other higher granularity can be a city or county with multiple regions. Another higher granularity can be a state or the like. In addition, particle sizes smaller than zip codes may be determined if location information is available for such granularities. For example, if GPS location information is available, granularities such as road groups or a single road can be determined.

According to one embodiment, a minimum geolocation parameter is determined that has statistical significance for estimating the effect of online behavior on offline sales. The minimum geolocation parameter is determined from a plurality of granularity of users' geolocation parameters. The minimum geolocation parameter is used to correlate online behavior with offline sales for impact estimation. The determination of the minimum geolocation parameter is based on the statistical significance of the granularity of the geolocation parameters. The statistical significance may depend on determining the minimum sample or data points needed to avoid or minimize the probability of being associated with the statistical validity of the data and not detecting the actual effect in the data. The minimum geographic location parameter with statistical significance may be the granularity of the geographic location parameters for users, and in this case, sufficient online behavior data and offline data sufficient to detect or estimate the effect of online behavior on offline sales, Lt; / RTI >

For example, if online activity for a product is captured for all users of a particular postal code, but there is a minimum of offline sales data for the same product of that postal code, Can not be reliably determined. However, offline sales data may be available for an area containing multiple postal codes. These areas are used to correlate online and offline data for impact estimation It can be the minimum geolocation parameter.

Other variables besides the amount of online and offline data are used to determine the minimum geolocation parameter. These variables include the type of product or brand, the product's purchase cycle, IP penetration (e.g., granularity level for geo-location parameters that can be determined by reverse IP lookup), site visit frequency, tracking / capture within online data Seasonality factors such as conversion rate of the resulting behavioral result, retail store / shop density, website traffic, and time of year, holidays, and the like.

In addition, the minimum geolocation parameter is dynamic. This can change over time as the variables change.

2. Method

FIG. 1 illustrates a method 100 for determining a minimum geolocation parameter in accordance with one embodiment. At step 101, an online action for the product is captured. A product can be a single product or a group of products. For example, a product can be a facial cream or a skin care product sold under the same brand. A product can be a consumer goods or service. An example of a service is a cellular telephone service for sale. Online actions include events monitored over the Internet, such as website traffic on a product's website, click-through rates for online ads or coupons, and the like. The online action is stored, for example, in the DB.

At step 102, offline data for the product is captured. This includes any in-store sales of the product, use of the coupon for purchase of the product, or any event associated with the sale of the product, but not an online event. In addition, the offline data can be stored in the DB.

At step 103, the geolocation parameters for the online behavior are determined. This includes multiple granularity of geolocation parameters for users performing events that are captured as online actions. The minimum granularity may comprise a set of blocks in a city or may be a zip code. Larger granularities may include a collection of zip codes, a collection of cities, or a state-level granularity. The minimum granularity depends on geographic location parameter data available to the user. For example, reverse IP lookup is performed to determine the minimum granularity. Regional location data available for reverse IP lookups may vary by region. For populated areas, such as metropolitan areas, the user's IP address may be cross-referenced with a set of city blocks. For more populated regions, the user's IP address may be cross-referenced to a single postal code or larger area.

In step 104, the minimum granularity geolocation parameter is determined from the granularity of the geolocation parameters determined in step 103. [

At step 105, a determination is made as to whether the minimum granularity geolocation parameter determined at step 104 has a statistical significance to estimate the effect of online behavior on offline sales for the product. A number of variables and other variables are used to evaluate the statistical significance, including the amount of online behavior data and the amount of off-line data for the minimum granularity geolocation parameter determined from the data captured in steps 101 and 102. As discussed above, examples of other variables include the type of product or brand, the purchase cycle of the product, the IP penetration rate (e.g., granularity level for geographic location parameters that can be determined by reverse IP lookup), site visit frequency, Seasonal factors such as retailer / shop density, website traffic, and year-round time, holidays, and so on.

If it is determined in step 105 that the minimum granularity geolocation parameter does not have statistical significance, then in step 106 the geolocation parameter is aggregated and step 105 is repeated. Integration includes determining the next higher granularity geolocation parameter among the geolocation parameters determined in step 103. [ Different granularities, e.g., city block, zip code, region of postal codes, and stock may be predetermined. Beginning at the minimum granularity, the online behavior data can then be incorporated into higher granularity or the like. For example, all online behavior data is stored for a particular city block group. On-line behavior data is integrated for all the city block groups within the zip code to determine the online behavior data for the next highest granularity, such as zip codes containing multiple city block groups. This integration may include associating all of the online behavior data for each user in the city block groups with the zip code. The zip code can be used as an index to determine all online behavior data for the products in that zip code. The online behavior data for the postal code may then be used in step 105 to determine whether the postal code has statistical significance in estimating the effect of online behavior on offline sales for the product. This integration is repeated until the granularity of the geo-location parameter with statistical significance is determined.

In step 107, after the granularity of the geo-location parameters with statistical significance is determined, the granularity is used as the minimum granularity geo-location parameter to estimate the effect of online behavior on offline sales for the product.

At step 108, Marketing Return On Investment (MROI) econometric modeling can be used to estimate the impact of online behavior on offline sales of the product. The inputs to the modeling include on-line behavior data and offline data for the minimum geolocation parameter data. Other inputs for modeling may include information about marketing campaigns, local competitor behavior, demographics, and so on. The MROI model may include historical data of marketing on the resulting impact of marketing.

The method (100) can be used for online marketing Can be used for optimization. For example, if an online activity is determined to affect off-line sales for a particular zip code, the website may be updated to better meet the requirements for that zip code Can be optimized. This may entail providing different web sites for different zip codes. For example, a website for a single zip code can be designed to market a luxury passenger car brand, which determines that the online behavior for that zip code affects the sale of luxury cars at local dealerships It is because. Another zip code website can be designed to promote hybrid vehicles because online and offline sales data indicate that customers are interested in green products. Targeted region-by-region online marketing campaigns can be generated in response to MROI modeling, which can improve product sales.

The method 100 is not performed merely for a snapshot of online and offline data. Instead, on-line and off-line data are continuously and periodically captured and used to continuously or periodically determine the minimum geolocation parameters. As a result, the minimum geolocation parameter for a product can change over time due to changes in online and offline data. Moreover, the changing minimum geolocation parameter allows the on-line behavior to continuously or periodically estimate the impact of online activity on offline sales. As a result, marketing campaigns or Other business conventions can be optimized periodically based on changes in online and offline user behavior.

3. System

FIG. 2 illustrates a system 200 in accordance with one embodiment. A plurality of users 201a-n access the website 211 via the Internet. For each captured online action of the user, an IP address is obtained from the web site request. The reverse IP lookup module 202 determines geolocation parameters for each user. This can be done with a look-up table that correlates IP addresses with zip codes or regions. Modules, such as those used herein, may include software executed by a processor or other processing circuitry. Instead of the reverse IP lookup, GPS or other systems and techniques may be used to determine the geographic location parameters of the user. A reverse IP lookup, GPS, or other geolocation capture system is used to determine the minimum granularity for each of the user's captured online actions. This information can be integrated to determine the captured online behavior of users for higher granularities.

The geographical location parameters for each user are stored in the online data DB 203. [ The online data DB 203 stores corresponding geographical location parameters together with the online actions captured for each user. The online behavior data is used as an input to the dynamic localization module 205 along with corresponding geographic location parameters. Examples of fields in the DB 203 may be product, geographic location parameter granularity, captured online behavior, and user information, if available.

In addition to the online behavior data, the system 200 captures offline data. Retail stores 210a-x capture the corresponding geographic location parameters for offline sales, such as offline sales data and store locations where sales are made. The offline data DB 204 stores offline sales data and corresponding geographic location parameters. Examples of fields in the DB 204 may be product, geographic location parameter granularity, offline data, and user information, if available.

Online and offline data can be captured and stored continuously or periodically for many products. If the effect of online behavior on offline sales for a particular product needs to be determined, the online and offline data for that product is sent to the dynamic localization module 205. For example, dynamic localization module 205 retrieves online and offline product data from DBs 203 and 204. Next, using the data and other variables 211, the dynamic localization module 205 determines a minimum geographic location parameter having statistical significance for estimating the effect of online behavior on offline sales for the product do. The dynamic localization module 205 may use the method 100 to determine a minimum geolocation parameter with statistical significance. Variables 211 for this determination may include the amount of online and offline data at a particular granularity, the type of product or brand, the product's purchase cycle, the IP penetration rate (e.g., geographic location parameters that may be determined by the reverse IP lookup) Seasonality factors such as the frequency of site visits, the frequency of site visits, the conversion rate of behavioral results tracked / captured within online data, retail / store density, website traffic, and year-round and holiday periods.

One example of how the variables 211 are used to calculate the statistical significance include one or more thresholds. For example, one threshold is set for the amount of online and offline data needed to estimate the effect of online behavior on offline sales. Such a threshold value may be determined depending on the product. Once sufficient online and offline data is available and the granularity satisfying the threshold is identified, other parameters may be used to determine whether the granularity has statistical significance. In another example, variables are used for threshold adjustment. For example, for a product that is determined to increase in sales at certain times of the year (ie, seasonal factors) based on historical sales figures, a particular granularity may be identified as having statistical significance during the year More offline sales data may be required. Other estimates may include determining a score for each variable and selecting the granularity based on the scores.

The integration module 206 integrates all on-line and off-line product data for the minimum geolocation parameter. The online and offline product data is retrieved, for example, from DBs 203 and 204. This may include retrieving all corresponding on-line and off-line data regarding the minimum geolocation parameter using the product and minimum geolocation parameters. If the online and offline data is stored in DBs 203 and 204 with a smaller granularity (e.g., zip code) and the minimum geolocation parameter is a larger granularity (e.g., weeks) Retrieve and integrate all online and offline product data for each postal code within the state.

The MROI model 207 uses aggregated data and MROI data 212, such as information about marketing campaigns, local competitor behavior, demographics, and the like. The MROI model 207 generates impact data 208 that includes an estimate of how online behavior affects offline sales for a product. The system 200 may be used to periodically or continuously determine the impact data 208.

The MROI model (207) is an econometric model that separates the effect of online behavior on off-line sales with respect to the minimum geographic location parameter. The MROI model 207 may include years of sales and marketing data to increase accuracy and harmonize variables such as seasonality. Marketing response curves are generated by calculating the relationship between marketing and sales at different investment levels to obtain average and marginal ROI from historical sales and marketing data. The financial model converts these response curves into net profits. By redistributing money from one driver to another driver, you can create a more optimal mix of marketing spending.

3 shows the online data 301 and the off-line data 302 (e.g., data from the integration module 206 shown in FIG. 2) provided in the MROI model 207. FIG. The minimum geolocation parameter 303, which is the data determined by the dynamic localization module 205 shown in FIG. 2, is also provided to the MROI model 207. In one example, based on historical sales data in the MROI model 207, a response curve for the minimum geographic location parameter is generated that represents online behavior for the product as a function of investment and offline sales. FIG. 4 shows an example of a response curve according to an investment in which the x-axis is an investment amount equivalent to dollars consumed in online marketing and the y-axis is offline sales. This curve shows an increase in online behavior, such as website traffic, which is correlated with an increase in offline sales and investment. The MROI model 207 may generate response curves for other marketing such as TV / radio marketing.

The financial model 304 shown in FIG. 4 converts the response curves into net gains based on the historical data. Business practices can then be modified based on the output of the models, as shown at 305. In addition, corrections can be customized for specific regions based on the minimum geolocation parameter. This may include increasing or decreasing investment in different types of marketing in different regions. In addition, web sites can be geographically modified to match different effects on offline sales to different regions.

4. Computer readable medium

FIG. 5 illustrates a computer system 500 that may be used with the embodiments described herein. Computer system 500 represents a generic platform that includes components that may reside in a server or other computer system. Computer system 500 may be used as a platform for executing one or more methods, functions, and other steps described herein. These steps may be implemented as software stored on one or more computer readable media.

Computer system 500 includes a processor 502 that can implement or execute software instructions that perform some or all of the methods, functions, and other steps described herein. The instructions and data from the processor 502 are communicated via the communication bus 504. The computer system 500 includes a main memory 506, such as RAM (Random Access Memory), in which software and data for the processor 502 may reside during execution time, and a memory 502 that may be non- Lt; RTI ID = 0.0 > 508 < / RTI > Memory and data storage devices are examples of computer readable media.

The computer system 500 may include one or more I / O devices 510, such as a keyboard, a mouse, a display, The computer system 500 may include a network interface 512 for connecting to a network. It will be apparent to one of ordinary skill in the art that other known electronic components may be added to or substituted for the computer system 500.

One or more of the steps of the methods described herein and the other steps described herein and one or more components of the systems described herein may be implemented in a computer, such as a memory and / May be embodied as computer code stored on a readable medium and executed in a computer system, e.g., by a processor, an application specific integrated circuit (ASIC), or other controller. The code may reside as a software program (s) comprised of program instructions of source code, object code, executable code or other formats. Examples of computer readable media include conventional computer system RAM, Read Only Memory (ROM), Erasable, Programmable ROM (EPROM), Electrically Erasable, Programmable ROM (EEPROM), hard drive and flash memory.

Although the embodiments have been described with reference to examples, those skilled in the art will be able to make various changes to the embodiments described without departing from the scope of the claimed embodiments.

Figure 1 illustrates a method for determining a minimum geolocation parameter according to one embodiment.

2 illustrates a system for determining minimum geolocation parameters and estimating the impact of online behavior on off-line sales according to one embodiment.

3 illustrates the use of MROI modeling to modify business practices based on an estimate of the impact of online behavior on off-line sales, in accordance with one embodiment.

Figure 4 shows the response curve generated by MROI modeling according to one embodiment.

5 illustrates a computer system that may be used for a method and system in accordance with one embodiment.

Claims (20)

A computer system for correlating offline sales and online behavior for a product, A data storage device for storing offline sales data and online behavior data for the product; A dynamic localization module implemented by a processor, the dynamic localization module comprising: Determining an online behavior associated with the product of the users, Determining granularity of geographic location parameters for the online behavior of users, Determining a minimum granularity of the granularity of the geographic location parameter, Determining whether the minimum granularity has statistical significance for estimating the effect of the online behavior on offline sales of the product based on the online behavior data and the offline sales data and one or more parameters associated with the on- Respectively, In response to determining that the minimum granularity has statistical significance, using the minimum granularity as a minimum geographic location parameter, Determining whether the next higher granularity of the granularities has statistical significance, in response to determining that the minimum granularity has no statistical significance, repeatedly determining until the granularity having statistical significance is determined, As the minimum geolocation parameter; And An integration module, implemented by the processor, for integrating the stored online behavior data and offline sales data based on the minimum geolocation parameter, ≪ / RTI > The method according to claim 1, Includes a Marketing Return on Investment (MROI) model, Wherein the consolidated data is entered into the MROI model to determine the effect of the online behavior on the offline sales of the product with respect to the minimum geolocation parameter. 3. The method of claim 2, Wherein the dynamic localization module is operable to derive the online behavior data required to avoid or minimize the probability of not detecting the actual impact of the online behavior on the offline sales with respect to the minimum granularity and the minimum number of samples pertaining to the off- Wherein the minimum granularity has a statistical significance. The method of claim 3, Wherein one of the granularities is based on at least one of an online act data for each granularity and an amount of offline sales data, a product type, a brand type, a purchase cycle of the product and a seasonality factor, Selected as the location parameter. A method of correlating offline sales with online behavior using a computer system, The computer system includes a network interface, a data storage device, a dynamic localization module, and an integration module, The method comprises: Receiving online behavior data captured from web sites and offline sales data for products captured from store sales by the network interface; Storing the received online behavior data and offline sales data by the data storage device; Determining granularity of geographic location parameters for users' online behavior by the dynamic localization module; Determining, by the dynamic localization module, a minimum granularity of the granularity of the geographic location parameter; Determining whether the minimum granularity has statistical significance for estimating the effect of the online behavior on offline sales of the product based on the online behavior data and the offline sales data and one or more parameters associated with the on- Determining by the dynamic localization module; Using the minimum granularity as a minimum geographic location parameter by the dynamic localization module in response to determining that the minimum granularity has statistical significance; Determining, in response to the determination that the minimum granularity has no statistical significance, repeatedly determining by the dynamic localization module until a granularity with statistical significance is determined whether the next higher granularity of the granularities has statistical significance, Selecting the granularity having statistical significance as the minimum geographic location parameter by the dynamic localization module; And Integrating data about the online behavior and data about offline sales using the minimum geolocation parameter by the integration module ≪ / RTI > 6. The method of claim 5, From the minimum number of samples for the on-line activity data and the off-line sales data necessary to avoid or minimize the probability of not detecting the actual effect of the online behavior on the offline sales about the minimum granularity, ≪ / RTI > further comprising the step of determining by the dynamic localization module. 6. The method of claim 5, And integrating the stored online behavior data and offline sales data by the integration module based on the minimum geolocation parameter. At least one computer readable medium storing computer code, The computer code, when executed on one or more computer systems, Determine the online behavior associated with the user ' s product, Determining granularity of geographic location parameters for the online behavior of users, Determining a minimum granularity of the granularity of the geographic location parameter, Determine whether the minimum granularity has statistical significance for estimating the effect of the online behavior on offline sales of the product based on the one or more parameters associated with the online activity and offline sales, In response to determining that the minimum granularity has statistical significance, using the minimum granularity as a minimum geographic location parameter, Determining whether the next higher granularity of the granularities has statistical significance, in response to determining that the minimum granularity has no statistical significance, repeatedly determining until the granularity having statistical significance is determined, As the minimum geolocation parameter, And integrates data about the online behavior and data about offline sales using the minimum geolocation parameter. 9. The method of claim 8, From the minimum number of samples for the on-line activity data and the off-line sales data necessary to avoid or minimize the probability of not detecting the actual effect of the online behavior on the offline sales about the minimum granularity, At least one computer readable medium. delete delete 9. The method of claim 8, In order to determine whether the minimum grain size has statistical significance, Identifying one or more variables operable to be used to characterize statistical significance of the geographic location parameter for users of the online behavior, And determine, using the at least one variable, whether the minimum geolocation parameter has statistical significance. 13. The method of claim 12, Wherein the one or more variables include an amount of online behavior data and offline sales data for each granularity, a product type, a brand type, a purchase cycle of the product, and a seasonality factor. 13. The method of claim 12, At least one computer readable medium in which the minimum geolocation parameter is determined over time as one or more of the online and offline product data and the variable change over time. 9. The method of claim 8, The code includes: Inputting an online behavior associated with the product of the users with respect to the minimum geolocation parameter into a marketing return on investment (MROI) model, Inputting the offline sales data of the product related to the minimum geolocation parameter into the MROI model, And estimates the effect of the online behavior on the offline sales of the product with respect to the minimum geolocation parameter. 16. The method of claim 15, The code includes: And modify the business practices associated with the product in response to the estimated impact. 16. The method of claim 15, The code includes: And modify the website related to the product associated with the minimum geolocation parameter in response to the estimated impact. 9. The method of claim 8, The code includes: And performs a reverse Internet Protocol lookup on the events captured for the online action to determine the geographic location parameter. 9. The method of claim 8, Wherein the user geographic location parameter for the users comprises a geographic location for each of the users. 20. The method of claim 19, Wherein the geographic location parameter comprises at least one of GPS coordinates, a group of city blocks, a zip code and a region containing a number of neighboring zip codes.

Images